Traditional systems allow users to associate comments with specific media assets or segments. These systems however fail to account for the lack of standardization between different comments users have. The traditional systems also only consider when (e.g., time stamps) the comments were made but fail to identify the reasons behind the user comments and specifically to whom or what in the media asset they pertain. Accordingly, the traditional systems fail to provide accurate search results and recommendations as such search and recommendation engines operate with standardized terms and look for deeper connections within media assets.